Ethernet ring protection (ERP) is a resiliency mechanism in which may be used to fulfill high availability requirements while at the same time guaranteeing loop free topologies in Ethernet networks. ERP may for example be applied in Carrier Ethernet networks when striving at protection switching times which are equivalent to those of conventional Time Division Multiplexing (TDM) networks, i.e., in the range of 50 ms or less. Examples of ERP protocols are ITU-T G.8032, the Spanning Tree Protocol (STP) or STP variants, e.g., the Rapid Spanning Tree Protocol (RSTP) or the Multiple Spanning Tree Protocol (MSTP).
A G.8032 Ethernet protection ring has one ring protection link (RPL) which is blocked in normal operation, thus avoiding loops. In case of a failure of a link or port in the Ethernet protection ring, signal failure (SF) messages are transmitted between ring nodes of the Ethernet protection ring multicast to inform ring nodes maintaining the RPL of the failure condition. These ring nodes may then unblock the RPL, thereby forming a new network topology. Details of the operation of a G.8032 ERP system can be found in ITU-T G.8032 and IEEE 802.1D.
Ethernet networks may be based on various types of link types, including for example microwave links and other types of wireless links. Some of such link types may offer the possibility of bandwidth adaptation. For example, the bandwidth of a microwave link may be adapted by adapting a modulation scheme of the link, enabling microwave nodes to choose the highest order modulation depending on radio conditions. At good radio conditions, with high signal to noise ratios, higher order modulation may be used in order to gain high throughput. Instead of accepting higher bit error rates during times of lower signal to noise ratios, e.g., due to channel fading, a lower order modulation may be used, thereby reducing the bit error rate at the expense of lowering the transmission rate of the link.
Generally, it is also desirable to implement resilience in Ethernet networks which are at least in part based on links offering the possibility of bandwidth adaptation. Accordingly, there is need for techniques which allow for efficiently using bandwidth adaptation in an Ethernet ring protection system.